scholarly journals Ordinal SuStaIn: Subtype and Stage Inference for Clinical Scores, Visual Ratings, and Other Ordinal Data

2021 ◽  
Vol 4 ◽  
Author(s):  
Alexandra L. Young ◽  
Jacob W. Vogel ◽  
Leon M. Aksman ◽  
Peter A. Wijeratne ◽  
Arman Eshaghi ◽  
...  

Subtype and Stage Inference (SuStaIn) is an unsupervised learning algorithm that uniquely enables the identification of subgroups of individuals with distinct pseudo-temporal disease progression patterns from cross-sectional datasets. SuStaIn has been used to identify data-driven subgroups and perform patient stratification in neurodegenerative diseases and in lung diseases from continuous biomarker measurements predominantly obtained from imaging. However, the SuStaIn algorithm is not currently applicable to discrete ordinal data, such as visual ratings of images, neuropathological ratings, and clinical and neuropsychological test scores, restricting the applicability of SuStaIn to a narrower range of settings. Here we propose ‘Ordinal SuStaIn’, an ordinal version of the SuStaIn algorithm that uses a scored events model of disease progression to enable the application of SuStaIn to ordinal data. We demonstrate the validity of Ordinal SuStaIn by benchmarking the performance of the algorithm on simulated data. We further demonstrate that Ordinal SuStaIn out-performs the existing continuous version of SuStaIn (Z-score SuStaIn) on discrete scored data, providing much more accurate subtype progression patterns, better subtyping and staging of individuals, and accurate uncertainty estimates. We then apply Ordinal SuStaIn to six different sub-scales of the Clinical Dementia Rating scale (CDR) using data from the Alzheimer’s disease Neuroimaging Initiative (ADNI) study to identify individuals with distinct patterns of functional decline. Using data from 819 ADNI1 participants we identified three distinct CDR subtype progression patterns, which were independently verified using data from 790 ADNI2 participants. Our results provide insight into patterns of decline in daily activities in Alzheimer’s disease and a mechanism for stratifying individuals into groups with difficulties in different domains. Ordinal SuStaIn is broadly applicable across different types of ratings data, including visual ratings from imaging, neuropathological ratings and clinical or behavioural ratings data.

2013 ◽  
Vol 7 (2) ◽  
pp. 181-189 ◽  
Author(s):  
Margarida Sobral ◽  
Constança Paúl

ABSTRACT Education and participation in leisure activities appear to be highly relevant variables in Alzheimer's disease (AD) and usually form the basis of the Cognitive Reserve construct. Objective: [A] To determine the association between education, cognitive and functional ability of AD patients; [B] To determine the association between participation in leisure activities and cognitive and functional ability of AD patients; [C] To evaluate the association of education and participation in leisure activities in the course of AD. Methods: Functional and neuropsychological abilities of 120 outpatients with probable AD were evaluated at baseline, at 36 and 54 months. Data collected at baseline included socio-demographics, clinical variables, education and frequency of participation in leisure activities throughout life. All participants and/or caregivers answered the questionnaire, "Participation in leisure activities throughout life" while patients completed the MMSE, the Clinical Dementia Rating scale, neuropsychological tests from the Lisbon Screening for Dementia Assessment, Barthel Index and Lawton and Brody's Index. Results: AD patients with higher levels of education achieved better results on cognitive tests. The participants with higher participation in leisure activities exhibited better results on cognitive and functional tests than those with lower participation. The disease progression was linear and progressed similarly regardless of the level of education of participants. However, the results suggest a slower disease progression in patients with a higher level of participation in leisure activities throughout their lives. Conclusion: AD patients with high education and high participation in leisure activities may benefit from a slower cognitive and functional decline after diagnosis of AD.


2019 ◽  
Vol 34 (5) ◽  
pp. 314-321
Author(s):  
Miwako Takahashi ◽  
Tomoko Tada ◽  
Tomomi Nakamura ◽  
Keitaro Koyama ◽  
Toshimitsu Momose

This study aimed to assess efficacy and limitations of regional cerebral blood flow imaging using single-photon emission computed tomography (rCBF-SPECT) in the diagnosis of Alzheimer’s disease (AD) with amyloid-positron emission tomography (amyloid-PET). Thirteen patients, who underwent both rCBF-SPECT and amyloid-PET after clinical diagnosis of AD or mild cognitive impairment, were retrospectively identified. The rCBF-SPECTs were classified into 4 grades, from typical AD pattern to no AD pattern of hypoperfusion; amyloid-beta (Aβ) positivity was assessed by amyloid-PET. Four patients were categorized into a typical AD pattern on rCBF-SPECT, and all were Aβ+. The other 9 patients did not exhibit a typical AD pattern; however, 4 were Aβ+. The Mini-Mental State Examination score and Clinical Dementia Rating scale were not significantly different between Aβ+ and Aβ– patients. A typical AD pattern on rCBF-SPECT can reflect Aβ+; however, if not, rCBF-SPECT has a limitation to predict amyloid pathology.


2011 ◽  
Vol 24 (2) ◽  
pp. 197-204 ◽  
Author(s):  
Alessandro Sona ◽  
Ping Zhang ◽  
David Ames ◽  
Ashley I. Bush ◽  
Nicola T. Lautenschlager ◽  
...  

ABSTRACTBackground: The AIBL study, which commenced in November 2006, is a two-center prospective study of a cohort of 1112 volunteers aged 60+. The cohort includes 211 patients meeting NINCDS-ADRDA criteria for Alzheimer's disease (AD) (180 probable and 31 possible). We aimed to identify factors associated with rapid cognitive decline over 18 months in this cohort of AD patients.Methods: We defined rapid cognitive decline as a drop of 6 points or more on the Mini-Mental State Examination (MMSE) between baseline and 18-month follow-up. Analyses were also conducted with a threshold of 4, 5, 7 and 8 points, as well as with and without subjects who had died or were too severely affected to be interviewed at 18 months and after, both including and excluding subjects whose AD diagnosis was “possible” AD. We sought correlations between rapid cognitive decline and demographic, clinical and biological variables.Results: Of the 211 AD patients recruited at baseline, we had available data for 156 (73.9%) patients at 18 months. Fifty-one patients were considered rapid cognitive decliners (32.7%). A higher Clinical Dementia Rating scale (CDR) and higher CDR “sum of boxes” score at baseline were the major predictors of rapid cognitive decline in this population. Furthermore, using logistic regression model analysis, patients treated with a cholinesterase inhibitor (CheI) had a higher risk of being rapid cognitive decliners, as did males and those of younger age.Conclusions: Almost one third of patients satisfying established research criteria for AD experienced rapid cognitive decline. Worse baseline functional and cognitive status and treatment with a CheI were the major factors associated with rapid cognitive decline over 18 months in this population.


2016 ◽  
Vol 41 (5-6) ◽  
pp. 314-323 ◽  
Author(s):  
Fabricio Ferreira de Oliveira ◽  
Elizabeth Suchi Chen ◽  
Marilia Cardoso Smith ◽  
Paulo Henrique Ferreira Bertolucci

Background: Midlife hypertension followed by late life hypotension resulting from neurodegeneration increases amyloidogenesis and tauopathy. Methods: Consecutive outpatients with late-onset Alzheimer's disease (AD) at various stages and their respective caregivers were assessed for score variations in 1 year of tests assessing caregiver burden, functionality and cognition according to blood pressure (BP) variations and APOE haplotypes, while also taking into account differential effects of angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, β-blockers, calcium channel blockers, diuretics, or no antihypertensive medication on score changes. The diagnosis and treatment of arterial hypertension followed the JNC 7 report. Results: Variations in systolic BP (-11.76 ± 17.1 mm Hg), diastolic BP (-4.92 ± 10.3 mm Hg) and pulse pressure (-6.84 ± 12.6 mm Hg) were significant after 1 year (n = 191; ρ < 0.01). For APOE4+ carriers, rises in systolic or diastolic BP improved Clinical Dementia Rating Scale Sum of Boxes scores (ρ < 0.04), with marginally significant improvements in Mini-Mental State Examination scores resulting from risen systolic (ρ = 0.069) or diastolic BP (ρ = 0.079), and in basic independence only regarding risen diastolic BP (ρ = 0.055). APOE4- carriers resisted any functional or cognitive effects of BP variations. No differences were found regarding any antihypertensive class for variations in BP or any test scores, regardless of APOE haplotypes. Conclusions: Targeting mild BP elevations brings better functional and cognitive results for APOE4+ carriers with AD.


2019 ◽  
Vol 48 (Supplement_3) ◽  
pp. iii1-iii16
Author(s):  
Claire Murphy ◽  
Adam H Dyer ◽  
Brian Lawlor ◽  
Sean P Kennelly

Abstract Background Prescription of Potentially Inappropriate Medications (PIMs) is common in older adults and is associated with adverse drug events, hospitalisation and mortality. Less well described are the patterns and predictors of PIM usage in patients with Alzheimer’s Disease (AD), a patient group who may be particularly vulnerable to polypharmacy and medication associated adverse events. Methods Secondary analysis of the NILVAD trial, an international phase three trial of Nilvadipine in mild/moderate AD. The v2 STOPP/START criteria were individually applied by a physician to each participant’s medication list and cross-reference with their medical history to identify PIM usage. Predictors of PIM usage were modelled using binary logistic regression. Results Five-hundred and ten patients with AD were included (mean age 72.8 +/-8.3 years; 62% female). The median number of prescribed medications was 5 (IQR 3-7). Over half (55.5%) were prescribed at least one PIM, whilst a minority of patients (14.8%) were prescribed three or more PIMs. The most frequent PIMs were benzodiazepines >4 weeks without indication (n = 55), long-term Proton-Pump Inhibitor (PPI) use without appropriate indication (n = 49), use of non-steroidal analgesics without use of PPI (n=19) and antimuscarinic use in dementia (n= 18). On multivariate analysis, significant predictors of PIM use were higher total number of medications (p=0.001; OR 1.52; 1.36-1.59) in addition to greater AD severity, as rated using the Clinical Dementia Rating Scale Sum-of-Boxes (CDR-sb) (p=0.024; OR 1.18; 1.02-1.35). Conclusion The majority of older patients with AD were prescribed at least one PIM. Usage of PIMs was associated with greater number of medications and increased dementia severity. Particularly concerning is the potentially inappropriate use of benzodiazepines and anti-muscarinic agents in this population, given recent evidence for the adverse cognitive profile associated with these medications. De-prescribing and medication review interventions aimed particularly at patients with AD are warranted.


BMJ Open ◽  
2020 ◽  
Vol 10 (8) ◽  
pp. e036990 ◽  
Author(s):  
MengFei He ◽  
Li Sun ◽  
Wenhui Cao ◽  
Changhao Yin ◽  
Wenqiang Sun ◽  
...  

IntroductionNeurogranin is known to be significantly elevated in patients with Alzheimer’s disease (AD) and may be an effective clinical predictor of cognitive decline and neurodegeneration. Amnestic mild cognitive impairment (aMCI) is an intermediate disease state between normal cognitive ageing and dementia, the latter of which can easily revert to AD. There remains significant uncertainty regarding the conversion of aMCI to AD, and therefore, elucidating such progression is paramount to the field of cognitive neuroscience. In this protocol study, we therefore aim to investigate the changes in plasma neurogranin in the early stage of AD and the mechanism thereof regarding the cognitive progression towards AD.Methods and analysisIn this study, patients with aMCI and AD patients (n=70 each) will be recruited at the memory clinic of the Department of Neurology of Hongqi Hospital affiliated with the Mudanjiang Medical University of China. Healthy older controls (n=70) will also be recruited from the community. All subjects will undergo neuroimaging and neuropsychological evaluations in addition to blood collection at the first year and the third year. We hope to identify a new biomarker of cognitive decline associated with AD and characterise its behaviour throughout the progression of aMCI to AD. This work will reveal novel targets for the therapeutic prevention, diagnosis and treatment of AD. The primary outcome measures will be (1) neuropsychological evaluation, including Mini-Mental State Examination, Montreal Cognitive Assessment, Clinical Dementia Rating scale, Shape Trail Test-A&B, Auditory Verbal Learning Test-HuaShan version; (2) microstructural alterations and hippocampal features from MRI scans; and (3) neurogranin levels in the neuronal-derived exosomes from peripheral blood samples.Ethics and disseminationThe ethics committee of the Hongqi Hospital affiliated with the Mudanjiang Medical University of China has approved this study protocol. The results will be published in peer-reviewed journals and presented at national or international scientific conferences.Trial registration numberChiCTR2000029055.


Sign in / Sign up

Export Citation Format

Share Document